The RTX parking laptops of Nvidia seem to be completely thoughtful of disruption

The moment many years have been waiting for has come. Nvidia has been making graphics cards for decades WindowsDrive -PC ecosystem – now the company wants to control the whole with “superchips”, starting with the RTX Spark. The RTX Spark chips announced on the weekend at Computex Tech Expo in Taiwan combine uniform storage, RTX graphics and the new part: the N1CPU. Nvidia owns GPUss the entire world of AI processed in data centers. But now it is also used for locally operated AI. It’s just a taste of what’s going to come now, but these are the first Windows devices that actually meet the overstretched name “KI-PC”. The fake AI PC

Microsoft has been talking about the idea of “KI-PCs” since 2024, but it has never felt real. Certainly, the copilot+ PCs have integrated neuronal processing units (NPUs) and started with 16 GB of RAM, but they didn't have the power to perform large voice models locally as well as your phone. Instead of building the hype machine around AI, Microsoft's early promises about the next PC era felt empty. Although they still have to be tested and prices are not yet fixed, these new Nvidia laptops really look like real AI PCs. The combination of a uniform memory of up to 128 GB, an efficient arm-based CPU and the company's brand-typical RTX graphics cards gives a computer that looks like it outside the MacBook Pro. This competition is important because the MacBook Pro (or Mac desktops) was the only serious option for AI enthusiasts who wanted to run models of the base level locally. Nvidia opens its technology for the usual Windows partners like HP, Asus, Dell, Lenovo and others. However, Microsoft's own Surface Laptop Ultra could be the highlight of this new PC generation. It is a real MacBook-Pro alternative with a 15-inch Mini LED display and several connections in a seemingly similar package. Microsoft hasn’t been offering a performance-oriented surface device for years, and this is ahead of the macBook Pro Ultra, which will be launched later this year. I'd lie if I said I wasn't excited. One of the problems with Windows laptops that try to compete with MacBook Pro is the battery life. Since they are always dependent on separate graphics cards (normally from Nvidia), things like battery life and fan noise tend to break in. Only this year, with the release of the new Core-Ultra series 3 chips from Intel, we have seen laptops like the Dell XPS 14, which can bring GPU performance and battery life closer to the level of MacBook Pro. However, these configurations reach a maximum of 64 GB of memory. It's not just about more storage. It is also the fact that the graphics will be as powerful as a discrete RTX 5070. And it is also the software layer (known as CUDA) that allows developers to access the system's GPU core. Due to the use in data centers, Nvidia has built up a highly developed AI platform around CUDA. The transfer of this AI processing power to local PCs could mean an AI power that takes far more out of the hardware than current computers. Mac development has grown rapidly last year, but is still far behind the extent that Nvidia could reach with RTX Spark. A Revolution

Do not be fooled by the fact that prices are affordable. Some reports assume that the prices for high-end configurations of RTX Spark laptops will be over 4,000 US dollars. This should not be a surprise, because so much today costs a similarly configured MacBook Pro. It is no longer difficult to imagine a near future in which local AI models are routinely used for conclusions in a variety of work projects. Since agent models are becoming more user-friendly and efficient, their local use is increasingly preferred for data protection reasons. We already see a sudden increase in demand for the Mac Mini, which is faced with long delivery times, what Apple to the surprisingly fast introduction of AI. RTX Spark also looks at Mac Mini and several compact desktop PCs with small form factor (SFF) are on the way. These devices can start with lower performance configurations to reach a wider audience. In the press releases of Nvidia and Microsoft it is all said that the devices will have “up to” 128 GB of storage and “up to” a 20-core CPU. Nvidia wanted to convince with the announcement of its maximum performance, especially as a successor to the DGX Spark desktop from the beginning of the year. But these lower configurations could appeal to buyers who are not AI enthusiasts, such as gamers or developers – or people who do a little of both. Nvidia finally still markets gaming features for these RTX parking laptops. The extent to which they emphasise and advance these categories will determine how great the impact on Intel, AMD and Qualcomm will be. But at the moment there is reason to be in the highest alertness. The exciting thing is that these devices have hardware that has not yet been available in the Windows ecosystem, creating a completely new device class. This should be a recovery for the Windows PC as a whole. So or so, as I used to be a sceptic of the idea of the “KI-PC”, I have to say that I am beginning to buy the vision.